Guest Contributor
Xiaochin Yan

Xiaochin Claire Yan is a Policy Fellow in Education Studies at the Pacific Research Institute in San Francisco.[Yan index]

California Needs a New Model
Measuring student achievement...
[Xiaochin Yan] 7/22/04

The federal No Child Left Behind Act mandates that every child be proficient in math and reading by 2013–14. The way California has structured its targets for meeting the NCLB requirements almost guarantees that the state’s schools will come up short. A major problem is the way California reports student test scores.

The current method provides a partial and often inaccurate picture of how students are really doing and the results are often of little help to teachers. Test-score reports focus on average test scores and growth at the school level, overlooking the needs of the individual student. The Pacific Research Institute proposes a new model for measuring student achievement that addresses the shortcomings of the current system.

What is the problem with using average test scores? Take Billy, a hypothetical low-performing student at a high-performing school. Because the school’s average test scores are high, exceeding state and federal goals, there may be little pressure on the school to help Billy improve. In order to assist him and all low-performing students, a measurement model is needed that provides information on how well individual students are progressing toward subject-matter proficiency.

Tracking a student’s testing history over the years helps to map out a trajectory for improvement. Yearly individual improvement goals can then be established for each individual student to help them reach proficiency by graduation.

In a recent study, Putting Education to the Test: A Value-Added Model for California, the Pacific Research Institute proposes a statistical model that provides annual achievement growth information for each individual student to meet federal and state goals of subject-matter proficiency.

This model answers the question, “Given a student’s level of performance, how much does he need to grow each year to become proficient in the subject matter by the time he leaves school?”

That knowledge, particularly in math and English, would be a valuable tool for a teacher starting the school year with a classroom full of new and unfamiliar students. This tool will allow the teacher to see how much extra attention a particular student may need in order to achieve his or her personal improvement goal.

Because the model measures and projects how each individual student is progressing toward proficiency, it can be used to evaluate whether a particular education program has helped or hurt student achievement. It can also reveal which instructional practices are best able to move students toward subject-matter proficiency, and help identify strong teachers and assist weak teachers with additional training.

Given the federal goals, implementing an accurate assessment system is more important than ever for California. In 2003-04 the state set a target of 13.6 percent of student proficiency in English and 16 percent proficiency in mathematics. Next year, the targets will each increase by approximately 10 percent, and so on, for the next ten years until the NCLB deadline in 2013-14.

Schools, pressured to hit these growth targets, will be tempted to focus teaching and resources on those students just below the proficiency level because it is easier to get middle-performing students to pass the benchmark, thus allowing schools to hit their year-by-year targets. This ignores those students who are far below, or above, the proficiency level.

With the NCLB clock ticking, states should consider tools like the PRI model that provide the individual diagnostic information necessary to help each individual student become successful. CRO

§